[PATCH v4 02/13] file-posix, gluster: Handle zero block status hint better

[Eric Blake]

Although the previous patch to change 'bool want_zero' into a bitmask
made no semantic change, it is now time to differentiate.  When the
caller specifically wants to know what parts of the file read as zero,
we need to use lseek and actually reporting holes, rather than
short-circuiting and advertising full allocation.

This change will be utilized in later patches to let mirroring
optimize for the case when the destination already reads as zeroes.

Signed-off-by: Eric Blake <ebl...@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
---
 block/file-posix.c | 3 ++-
 block/gluster.c    | 2 +-
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/block/file-posix.c b/block/file-posix.c
index 805a1a2949b..ec95b748696 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -3289,7 +3289,8 @@ static int coroutine_fn 
raw_co_block_status(BlockDriverState *bs,
         return ret;
     }

-    if (mode != BDRV_WANT_PRECISE) {
+    if (!(mode & BDRV_WANT_ZERO)) {
+        /* There is no backing file - all bytes are allocated in this file.  */
         *pnum = bytes;
         *map = offset;
         *file = bs;
diff --git a/block/gluster.c b/block/gluster.c
index 1a2ef53e9b0..89abd40f313 100644
--- a/block/gluster.c
+++ b/block/gluster.c
@@ -1478,7 +1478,7 @@ static int coroutine_fn 
qemu_gluster_co_block_status(BlockDriverState *bs,
         return ret;
     }

-    if (mode != BDRV_WANT_PRECISE) {
+    if (!(mode & BDRV_WANT_ZERO)) {
         *pnum = bytes;
         *map = offset;
         *file = bs;
-- 
2.49.0